Brain stroke detection using neural networks: A deep learning approach for early diagnosis and prevention

This study investigates the effectiveness of three machine learning algorithms—Naive Bayes, Decision Tree, and Random Forest—for stroke detection based on patient data. We utilize a publicly available dataset from Kaggle that includes clinical features such as age, gender, hypertension, and heart disease history. Each model is evaluated on its Area under the Receiver Operating Characteristics (AUROC), accuracy, precision, recall, and F1-score to determine the best performer in predicting stroke risk.